Jouni Malinen [Tue, 2 Nov 2010 17:35:40 +0000 (19:35 +0200)]
P2P: Skip extra remain-on-channel if oper channel matches
There is no need to separately move to the correct channel
for transmitting an Action frame that is using the group
interface (i.e., source address is P2P Interface Address).
This removes extra latency from P2P Action frame operations
within a P2P group.
Helmut Schaa [Sun, 31 Oct 2010 19:36:43 +0000 (21:36 +0200)]
nl80211: Pass data frames from unknown STAs to hostapd
Pass data frames from unknown STAs to hostapd in order to reply with
a Deauthentication or Disassociation frame. This fixes compliance
with IEEE Std 802.11-2007, 11.3.
Furthermore, this does not cause a lot of overhead (at least with
mac80211 drivers) since mac80211 does not pass all data frames (but
at least from unauthenticated and unassociated STAs) to cooked monitor
interfaces.
Tested with rt2800pci on a MIPS board.
Signed-off-by: Helmut Schaa <helmut.schaa@googlemail.com>
Dmitry Shmidt [Sun, 31 Oct 2010 10:38:49 +0000 (12:38 +0200)]
Add WPA_IGNORE_CONFIG_ERRORS option to continue in case of bad config
This is an option to continue with wpa_supplicant and hostapd even if
config file has errors. The problem is that these daemons are the best
"candidates" for the config change, so if they can not start because
config file was let's say corrupted, you can not fix it easily.
Jouni Malinen [Sun, 31 Oct 2010 10:27:37 +0000 (12:27 +0200)]
P2P: Fix random channel selection to use os_get_random
This could have been using an uninitialized variable instead of
purposefully random value in picking up the channel. There is not
much of a difference for the use case here, but anyway, we may
as well do what was initially planned here and.
Jouni Malinen [Sun, 31 Oct 2010 10:17:30 +0000 (12:17 +0200)]
Fix copy-paste bug in p2p_serv_disc_resp handling
The destination address for p2p_serv_disc_resp must always be set
to an individual MAC address unlike p2p_serv_disc_req which may use
00:00:00:00:00:00 to indicate wildcard query. As such, we should not
try to check for this special case here.
Jouni Malinen [Wed, 27 Oct 2010 16:36:10 +0000 (19:36 +0300)]
P2P: Do not re-send PD Request for join-a-group after acked frame
We are not actually interested in the PD Response in join-a-group
case, so there is no point in trying to send PD Request until the
response is received. This avoids an extra PD getting started after
a join-a-group operation in some cases.
Jouni Malinen [Wed, 27 Oct 2010 08:26:46 +0000 (11:26 +0300)]
Fix memory leak in P2P build without MLME
The ieee80211_sta_free_hw_features() function is now used outside
the MLME code, so better make sure it gets included in the build
to avoid leaking memory.
Jouni Malinen [Tue, 26 Oct 2010 13:30:28 +0000 (16:30 +0300)]
hostapd: Set operstate UP when initializing AP mode
This is needed to avoid problems with other applications setting and
leaving the interface to IF_OPER_DORMANT state. In AP mode, the interface
is ready immediately after the keys are set, so we better make sure the
DORMANT state does not prevent normal operations after that.
Jouni Malinen [Mon, 25 Oct 2010 19:22:07 +0000 (22:22 +0300)]
WPS ER: Cache AP settings for APs that restart UPnP
This is needed to avoid issues with APs that restart their UPnP,
e.g., when ER reconfigures them. The previously known settings are
now cached and taken into use if an AP is detected to leave
(ssdp:byebye) and then return.
Jouni Malinen [Mon, 25 Oct 2010 18:41:10 +0000 (21:41 +0300)]
WPS ER: Only activate PBC mode on single AP
Verify that the UUID given to wps_er_pbc command is known and only
activate PBC mode on the matching AP. The UUID can be that of the
AP or the station/Enrollee.
Jouni Malinen [Mon, 25 Oct 2010 16:16:11 +0000 (19:16 +0300)]
P2P: Remove P2P group on driver resource becoming unavailable
Add a new driver event, EVENT_INTERFACE_UNAVAILABLE, for indicating
that the driver is not able to continue operating the virtual
interface in its current mode anymore, e.g., due to operating
channel for GO interface forced to a DFS channel by another virtual
interface.
When this happens for a P2P group interface, the P2P group will
be terminated and P2P-GROUP-REMOVED event shows the reason for
this as follows:
P2P-GROUP-REMOVED wlan0 GO reason=UNAVAILABLE
Jouni Malinen [Mon, 25 Oct 2010 15:56:49 +0000 (18:56 +0300)]
WPS: Fix wps_reg command documentation to require BSSID
Replacing the BSSID with 'any' has never worked with wps_reg
command, so better not claim it could work. In addition, reject
this is wpa_supplicant to make it clearer that the operation
fails.
Jouni Malinen [Mon, 25 Oct 2010 15:24:15 +0000 (18:24 +0300)]
P2P: Add mechanism for timing out idle groups
A new configuration parameter, p2p_group_idle, can now be used to set
idle timeout value for P2P groups in seconds (0 = no timeout). If set,
this values is used to remove P2P group (both GO and P2P client)
interfaces after the group has been idle (no clients/GO seen) for the
configuration duration.
The P2P-GROUP-REMOVED event is now indicating the reason for group
removal when known. For example:
P2P-GROUP-REMOVED wlan0 GO reason=REQUESTED
P2P-GROUP-REMOVED wlan1 client reason=IDLE
Jouni Malinen [Fri, 22 Oct 2010 14:43:23 +0000 (17:43 +0300)]
hostapd: Avoid crashing on station mode disassoc event
Some driver wrappers may end up indicating a disassociation or
deauthentication event without the address of the station, e.g.,
based on a previous non-AP mode event. Avoid crashing hostapd
by verifying that the require address parameter is available in
the event before processing it.
Jouni Malinen [Wed, 20 Oct 2010 16:41:20 +0000 (19:41 +0300)]
P2P: Delay send_action call if p2p_scan is in progress
In order to avoid confusing the driver with a new remain-on-channel
request, delay sending of a new Action frame if the driver indicates
Action frame RX diromg a scan.
Jouni Malinen [Wed, 20 Oct 2010 16:37:47 +0000 (19:37 +0300)]
P2P: Do not stop Listen state if it is on correct channel
This is needed to optimize response to GO Negotiation Request frames.
The extra remain-on-channel cancel followed by new remain-on-channel for
the same channel takes too much time with some driver/firmware
designs for the response to go out quickly enough to avoid peer
timing out while waiting for our response.
Jouni Malinen [Wed, 20 Oct 2010 16:30:55 +0000 (19:30 +0300)]
Fix crash on channel list changed event during driver init
The driver wrapper may indicate EVENT_CHANNEL_LIST_CHANGED before
completing init(). This would likely result in segfault when driver
channel list is being read as part of processing this event when
P2P is enabled.
Jouni Malinen [Wed, 20 Oct 2010 10:37:01 +0000 (13:37 +0300)]
WPS ER: Add wps_er_set_config to use local configuration
This command can be used to enroll a network based on a local
network configuration block instead of having to (re-)learn the
current AP settings with wps_er_learn.
Jouni Malinen [Tue, 19 Oct 2010 16:57:01 +0000 (19:57 +0300)]
WPS: Update Beacon/ProbeResp IE on wps_version_number changes
This test command is supposed to change the WPS version number in all
places immediately, so make sure that the IEs used in management
frames get updated immediately.
Jouni Malinen [Tue, 19 Oct 2010 15:38:49 +0000 (18:38 +0300)]
P2P: Scan all channels if GO is not found from common channels
This is not supposed to happen, but in theory, the GO may end up
changing its channels at some point. While it would be reasonable
to use one of the common channels, GO may be forced to use another
channel. The most likely reason for this would be if the P2P client
is enforcing only a single channel to be used (e.g., to avoid
multi-channel concurrent operations).
Jouni Malinen [Tue, 19 Oct 2010 15:10:28 +0000 (18:10 +0300)]
P2P: Track non-P2P members in the group and set Group Limit bit
The P2P group component is now tracking of associated stations
in the group and the Group Limit bit in the Group Capabilities
is updated based on whether there is room for new clients in
the group.
Jouni Malinen [Tue, 19 Oct 2010 14:17:04 +0000 (17:17 +0300)]
Fix max_stations to apply to associated STAs, not P2P peers
This driver capability limit was supposed to indicate limit on
how many stations can associate with us in the AP mode, not how
many P2P peers are stored in the peer table.
Jouni Malinen [Tue, 19 Oct 2010 09:47:33 +0000 (12:47 +0300)]
P2P: Skip GO Neg Conf ack failure workaround of send failures
The workaround to ignore no ctrl::ack received for GO Negotiation
Confirmation frame was only supposed to be used when the frame was
actually transmitted and just the ack was not received. However, due
to the way the driver failure on transmitting the frame were reported,
this ended up getting applied for all failures in sending the GO
Negotiation Confirmation frame.
Improve this by providing a mechanism to indicate whether send_action
operations fail locally before the frame was actually transmitted or
because of not receiving ack frame after having transmitted the frame.
Jouni Malinen [Mon, 18 Oct 2010 14:00:50 +0000 (17:00 +0300)]
P2P: Try to re-create group interface if pending one is not available
This is needed to work around cases where the pending interface gets
removed, e.g., due to failed GO Negotiation. In such case, we may still
accept new GO Negotiation to be completed, but the interface did not
get created without a separate p2p_connect command.
The current implementation is not ideal since the re-created interface
may, at least in theory, get incorrect interface addrees. Though, this
would likely require that the driver supported more than one P2P group
interface and that the previous one was already taken into use for
another group.
Jouni Malinen [Sun, 17 Oct 2010 18:36:04 +0000 (21:36 +0300)]
WPS: Do not drop subscriptions based on max queue length
UPnP event subscriptions are not supposed to be dropped based on
whether events can be delivered quickly enough. Leave dropping to
happen only based on failed deliveries to avoid issues with a burst
of events kicking out still active subscribers.
Jouni Malinen [Sun, 17 Oct 2010 18:24:12 +0000 (21:24 +0300)]
WPS UPnP: Throttle WLANEvent notifications to 5 per second
Do not send more than five Probe Request WLANEvent notifications
per second. Even though the limit should in theory apply to all
WLANEvents, it is better not to drop EAP notifications because
of Probe Request frames and really, the only real reason for
event bursts is Probe Request frames.
Jouni Malinen [Sun, 17 Oct 2010 17:58:58 +0000 (20:58 +0300)]
WPS: Limit Probe Request event queuing if subscriber may have left
Instead of queuing all events for a subscriber, only queue more
important events if delivery of event notifications starts failing.
This allows more time for the subscriber to recover since the maximum
queue length if not reached because of Probe Request frames only.
Instead of dropping the subscription on first failure, allow up to
10 failures before dropping. In addition, drop the callback URLs
one by one instead of full subscription if only one URL is failing.
Jouni Malinen [Sat, 16 Oct 2010 09:57:47 +0000 (12:57 +0300)]
WPS: Add virtual flags in Config Methods for WPS 2.0 if needed
This is a workaround for incorrect configuration (missing
virtual/physical identifier for config methods) for WPS 2.0 to
allow unmodified configuration from WPS 1.0 to be used while
enforcing compliant WPS 2.0 values.
Jouni Malinen [Fri, 15 Oct 2010 19:11:33 +0000 (22:11 +0300)]
Fix build with drivers that use driver_wext.c
If CONFIG_DRIVER_WEXT=y is not used in .config, but driver_wext.c
gets pull in to help another driver wrapper, rfkill code needs to
included to fix the build.
Jouni Malinen [Fri, 15 Oct 2010 15:56:36 +0000 (18:56 +0300)]
nl80211: Indicate channel list change events
Listen to regulatory event messages from kernel and convert them to
internal driver event notifications indicated that the channel list
may have changed.
P2P: Send AP mode WPS-FAIL event to parent interface
The AP operation with wpa_supplicant requires an additional callback
to get the needed event information from hostapd side so that
wpa_msg() can be called for wpa_s->parent if needed.
Jouni Malinen [Thu, 14 Oct 2010 17:49:54 +0000 (20:49 +0300)]
WPS: Add Config Error into WPS-FAIL events
This makes it easier to figure out what could have failed in the
WPS protocol and potentially provide more information for the
user on how to resolve the issue.
Jouni Malinen [Thu, 14 Oct 2010 11:24:56 +0000 (14:24 +0300)]
P2P: Reject multi-channel concurrent operations depending on driver
The driver wrapper can now indicate whether the driver supports
concurrent operations on multiple channels (e.g., infra STA connection
on 5 GHz channel 36 and P2P group on 2.4 GHz channel 1). If not,
P2P_CONNECT commands will be rejected if they would require
multi-channel concurrency.
The new failure codes for P2P_CONNECT:
FAIL-CHANNEL-UNAVAILABLE:
The requested/needed channel is not currently available (i.e., user has
an option of disconnecting another interface to make the channel
available).
FAIL-CHANNEL-UNSUPPORTED:
The request channel is not available for P2P.
Jouni Malinen [Tue, 12 Oct 2010 13:56:17 +0000 (16:56 +0300)]
P2P: Limit p2p_connect .. pbc join based on BSSID
Allow only the expected P2P Interface Address as the BSSID for
the AP to avoid selecting incorrect BSS should there be another
device that is advertising active PBC mode before the target
P2P GO does.
Jouni Malinen [Mon, 11 Oct 2010 13:07:49 +0000 (16:07 +0300)]
WPS: Workaround broken Credential encoding from some D-Link APs
At least D-Link DIR-600 and DIR-825 have been reported to include
an extra octet after the Network Key attribute within a Credential
attribute. This can happen at least when they are provisioning an
open network.
Add a workaround to detect this incorrectly encoded attribute and
to skip the extra octet when parsing such a Credential.
Johannes Berg [Sun, 10 Oct 2010 15:10:37 +0000 (18:10 +0300)]
nl80211: Use nl80211 for Probe Request/Response frames
The new nl80211 API means we don't need to use monitor interfaces. This
means that the P2P implementation now requires a kernel that has support
for generic management frame (not just Action frame) transmission.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Johannes Berg [Sun, 10 Oct 2010 14:52:13 +0000 (17:52 +0300)]
P2P: Fix remain-on-channel abort race
When the P2P state machine requests a remain- on-channel, there's a
potential race where it can then request a stop before the r-o-c has
actually started, in which case the stop will not be processed. Fix
that.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Johannes Berg [Sun, 10 Oct 2010 14:47:15 +0000 (17:47 +0300)]
nl80211: work around libnl bug
libnl has a bug, when binding more than two sockets and releasing one,
it will release the wrong address and then try to reuse it, which fails.
Therefore, we need to reimplement the socket address assignment logic
locally for libnl 1.1.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Paul Stewart [Sat, 9 Oct 2010 14:29:51 +0000 (17:29 +0300)]
dbus_new_handlers: Don't send NULL to dbus_message_new_error
The new DBus API helper function wpas_dbus_error_unknown_error
function can be called as a result of a failure within internal
getter calls, which will call this function with a NULL message
parameter. However, dbus_message_new_error looks very unkindly
(i.e, abort()) on a NULL message, so in this case, we should not
call it.
I've observed this course of events during a call to
wpas_dbus_getter_bss_wpa with a faileld parse of the IE parameter.
We got here through a call to fill_dict_with_properties which
explicitly calls getters with a NULL message parameter. Judging
from the way it is called, this could easily occur if an AP sends
out a malformed (or mis-received) probe response. I usually run
into this problem while driving through San Francisco, so I'm
exposed to any number of base stations along this path.
Daniel Kurtz [Sat, 9 Oct 2010 13:27:53 +0000 (16:27 +0300)]
dbus: Treat '' in SSIDs of Interface.Scan as a request for broadcast scan
This patch changes wpa_supplicant policy for handling '' in SSIDs field of
Interface.SSID DBus message. It treats '' (zero-length) SSID as a request
for a broadcast scan, instead of ignoring it.
This patch updates DBus API .Scan() logic per the test cases listed below:
1) Interface.Scan({'Type':'active', 'Channel':(2412, 20)})
Request: Active scan with only '' SSID (1 channel)
Should be: 1 broadcast ProbeRequest on specified channel
Previous: 1 broadcast ProbeRequest on specified channel
This Patch: 1 broadcast ProbeRequest on specified channel
2) Interface.Scan({'Type':'active', 'Channel':(2412, 20), 'SSIDs':['']})
Request: Active scan with only '' SSID (1 channel)
Should be: 1 broadcast ProbeRequest on specified channel
Previous: No ProbeRequests; passive scan results for specified channel
This Patch: FIXED: 1 broadcast ProbeRequest on specified channel
3) Interface.Scan({'Type':'active', 'Channel':(2412, 20), 'SSIDs':['MySSID']})
Request: Active scan with only non-'' SSIDs (1 channel)
Should be: 1 directed ProbeRequest for each SSID on specified channel,
no broadcast ProbeRequest
Previous: 1 directed ProbeRequest for each SSID on specified channel,
no broadcast ProbeRequest
This Patch: 1 directed ProbeRequest for each SSID on specified channel,
no broadcast ProbeRequest
4) Interface.Scan({'Type':'active', 'Channel':(2412, 20), 'SSIDs':['',
'MySSID']})
Request: Active scan with SSIDs, including 1 '' SSID (1 channel)
Should be: 1 broadcast ProbeRequest, 1 directed ProbeRequest for each
non-'' SSID on specified channel
Previous: 1 directed ProbeRequest for each non-'' SSID on specified
channel
This Patch: FIXED: 1 broadcast ProbeRequest, 1 directed ProbeRequest for
each non-'' SSID on specified channel
Daniel Kurtz [Sat, 9 Oct 2010 13:22:39 +0000 (16:22 +0300)]
dbus: Fix passive/active scans in some cases
Currently the DBus Interface.Scan API is counter-intuitive. It issues
ProbeRequests when doing passive scans when channels are specified, and
does not issue ProbeRequests for broadcast active scans.
This patch updates DBus API .Scan() logic per the test cases listed below:
1) Interface.Scan({'Type':'passive'})
Request: Passive scan (all channels)
Should be: No ProbeRequests; Passive Scan results for all channels
Previous: 1 ProbeRequest on all channels for both broadcast SSID and
selected network (scan_ssid=1)
This Patch: --No change--: 1 ProbeRequest on all channels for both
broadcast SSID and selected network (scan_ssid=1)
2) Interface.Scan({'Type':'passive', 'Channel':(2412, 20)})
Request: Passive scan (1 channel)
Should be: No ProbeRequests; Passive Scan results for 1 channel
(plus overlapping channels)
Previous: 1 broadcast ProbeRequest on specified channel
This Patch: --Fixed--: No ProbeRequests; Passive Scan results for 1
channel (plus overlapping channels)
3) Interface.Scan({'Type':'active'})
Request: Active scan with no SSIDs (all channels)
Should be: 1 broadcast ProbeRequest on all channels
Previous: No ProbeRequests; passive scan results for all channels
This Patch: --Fixed--: 1 broadcast ProbeRequest on all channels
4) Interface.Scan({'Type':'active', 'Channel':(2412, 20)})
Request: Active scan with no SSIDs (1 channel)
Should be: 1 broadcast ProbeRequest on specified channel
Previous: No ProbeRequests; Passive scan results for specified
channel (plus overlapping channels)
This Patch: --Fixed--: 1 broadcast ProbeRequest on specified channel